The impending aging crisis

One of the biggest economic issues that we’ll have to deal with in the near future is an aging population. This will result in a shrinking working age population that will have to finance the pensions of a growing retired population. Given the magnitude of this issue, it is quite pressing to know how aging will affect the economy.

It is no surprise that – everything else equal – aging will have detrimental effects on the economy. But everything else is usually not equal: people adjust to new situations and this must be accounted for. What adjustment mechanisms will people take? Furthermore, are pension reforms that are in progress in many developed countries (e.g. raising the retirement age) effective? Can these adjustment mechanisms and reforms help us avoid a crisis?

I attempt to answer these questions based on a paper by Vogel, Ludwig and Boersch-Supan (2013). The authors are primarily interested in three types of adjustments that can dampen the effects of population aging.

  1. Endogenous human capital formation. Population aging implies that labor will be a relatively more scarce resource. This will increase wages. Higher wages will then incentivize investment into human capital, i.e. they will increase returns to education.
  2. Investing abroad. Aging will also lead to lower returns on capital. It is a relatively standard result in economics that capital stock per capita increases with a lower population growth rate, and thus capital becomes more common*. This will then depress returns on capital. Given that this happens in old (developed) countries, it is possible for citizens of these countries to invest their savings abroad, in younger countries where returns are still high.
  3. Pension reform. This just refers to increasing the retirement age, which will lead to an extended work life. This can further incentivize investment in education, as wage income will be more important over the life cycle.

The first two of these adjustments are relatively hard – if not impossible – to control by policy makers. Nevertheless, we are interested in how they affect the economy, because this can shed some light on how wrong any model or argument that ignores these adjustments can be.

Let us first analyze the effects of these three adjustments on aggregate macro variables. Without any adjustments, aging clearly has a negative effect on GDP as the graph below shows. (Note: the graph shows detrended GDP per capita, this means long-term growth trends have been removed. If nothing changes, GDP per capita should stay constant at 100 in this graph.)

GDP with adjustments

The “exogenous, closed” dashed gray line refers to the situation in which no adjustment takes place. Human capital does not adjust (i.e. is exogenous), and agents can’t invest abroad (i.e. the economy’s closed). Once we take human capital adjustments into account, however, GDP will actually increase. Investment opportunities abroad are also beneficial.

From now on, let us stick with the most realistic scenario of an open economy with endogenous human capital adjustment. And let’s consider the third point above: raising the retirement age.

This will have even more positive effects on GDP per capita. A pension reform pushes out the retirement age and thus lengthens one’s work life. Labor income will then become more important over the life cycle. This will raise the importance of human capital even further. Higher human capital levels will then translate into higher productivity and GDP. The figure below compares the solid black line from the graph above (no pension reform) to the same scenario with pension reform.

GDP with pension reform, constant tau

There is one concern here, however. The pension system works the following way: working age people pay a contribution rate into some pension fund; all this income in the pension fund is then redistributed to retired people. Retired people’s pensions thus depend on the working age population’s size and wages.

With pressure on the pension funds from aging, two things can happen: either they raise the contribution rate (i.e. the pension “tax”), or they decrease the per person pay-outs to retirees. The scenarios presented above assumed the latter: a constant contribution rate, but decreasing pensions. What if we flip the coin and increase contribution rates to keep pensions constant? The situation is somewhat worse now.

GDP with pension reform, constant rho

We can see above that without raising the retirement age (solid black line), we’ll suffer losses to GDP per capita, despite the other two adjustment mechanisms (endogenous human capital and investing abroad). But with a higher retirement age (dashed gray), we can still avoid such losses.

So, we’ve talked about GDP, but in general we’re much more interested in the welfare of the population. This is hard to measure in reality, but not so in the model. The authors thus do welfare comparisons across various scenarios.

First, let’s talk about the current generation: those who are alive in 2010. Consider the figure below. The vertical axis measures consumption-equivalent variation (CEV): basically, there’s a gain in welfare if CEV is positive, and a loss if it’s negative.

In the left panel we see that if we keep pensions constant, then everybody but the oldest lose out. This is because we need to increase the contribution rate (the pension tax) in order to finance these pensions, and there are fewer and fewer young people to do that. Still, raising the retirement age (dashed black line) leads to lower losses than not doing so (solid black line).

Aging: welfare for current cohorts

In the right panel we see that if instead we keep the contribution rate constant, young people have gains, old people don’t see much change, but middle-aged people are much worse off than in the previous scenario. This is because a constant contribution rate inevitably leads to lower pensions (because fewer people are paying the contributions). Young people can still respond to this by investing more in education and preempting these negative changes. But middle-aged people cannot really invest in human capital anymore, so they’ll be hurt by the lower pensions.

Finally, what about future generations? The graph below is identical to the one above except that it looks at cohorts born between 2010 and 2050. Two things worth noting: (1) once again, increasing the retirement age is the best policy, it leads to the most (lowest) welfare gains (losses); (2) keeping pensions constant will inevitably lead to a welfare loss for all cohorts, keeping the contribution rate constant won’t.

Aging: welfare for future generations

So to sum up, what have we learned? We learned that the impending demographic changes will probably be accompanied by important behavioral changes in the population: people will invest more in education, and will be more likely to invest in foreign (= young country) assets. The former effect will be much more important, and it will help dampen the adverse effects of aging considerably.

We also learned that increasing the retirement age is a very desirable policy that will leave practically all cohorts better off relative to not doing anything. Taking it into account that life expectancy is growing, it is perhaps not so outrageous that people will have to work more.

Where we do have a conflict between generations is the decision on how to balance the pension fund’s budget: do we raise contribution rates or lower pension pay-outs? The former will undoubtedly leave currently middle-aged and old agents better off, while young people and future generations would prefer the latter.

To end this post, let me highlight two potential weaknesses in the model. It is assumed that agents retire at the retirement age and not before or after. In other words, the retirement age is exogenous. If, however, people will also change their behavior regarding retirement decisions (i.e. early or late retirements), then this could alter the results. Another issue may be whether in the real world people will adjust their human capital investments as nicely as they do in the model. If they don’t, then welfare losses are underestimated here.

Last thought: I never really understood why we can’t just have self-funded pensions. I.e. everybody’s just saving for themselves, and your pensions are your savings, period. We could make it obligatory to save this way, so people will need no government hand-outs once retired. This would avoid the whole “working age population is too small relative to the retired population” problem. Probably the biggest difficulty with such a system would be its implementation, or more precisely the transition to it from the current systems.


* = Simply consider that tomorrow’s capital stock is today’s aggregate savings. That is

K_{t+1} = N_t s_t,

where K is total capital stock, N_t is current population and s is savings. Divide both sides of this equation by tomorrow’s population to get

\frac{K_{t+1}}{N_{t+1}} = \frac{N_t}{N_{t+1}} s_t.

This gives us the result. The left-hand side is capital per capita, the right-hand side contains the inverse of population growth (N_t/N_{t+1}). So capital per capita is decreasing in population growth. Now, population aging is generally a result of low fertility (i.e. low population growth). Thus aging (~ lower population growth) leads to a higher capital stock per person.


2 thoughts on “The impending aging crisis

  1. There are a couple of problems with the idea of people “saving for themselves”:

    Where I live, contributing to a pension fund is not obligatory – at least, not yet.

    You will find that people are prone to dip into their savings as soon as a perceived “crisis” hits, believing that the situation will improve at “some point soon”.

    While the majority of people may understand the concept of saving, and should be able to do it, there are not that many workers out there who understand how, when, where and why to invest in a particular venture. The returns on savings are almost non-existent, while the returns on investment can be substantial.

    However, even having considered these drawbacks, when offered a position for the same Cost to Company, one with the option of a retirement annuity and one without, I will always choose the one without. For people with (even rudimentary) financial skills and an (basic) understanding of portfolio management, who also have the necessary discipline not to add this additional amount to their monthly household budget, the returns will almost ALWAYS be higher when investing for yourself. This is because a very large percentage of your retirement annuity contribution is actually going towards fees and very little towards actually funding your retirement.

    • I agree that a system where people save for themselves would present its own challenges. But it would be one solution to this whole aging problem.

      Regarding the issues you raise: I’m guessing if pensions were entirely self-financed then we would at a bare minimum need to (1) make it obligatory to save, (2) impossible to access your savings until you reach the retirement age, and (3) provide the vast majority of people who don’t know how to invest with some sort of standardized investment opportunity (e.g. pension funds could have a balanced “default” portfolio that is relatively low risk).

      In essence, how much freedom people should have in deciding where to invest in a save-for-yourself system is up to debate. But the point is that whatever you save is what you will get back when you retire (+ interest/returns). So the pension system’s budget wouldn’t have to be balanced across generations anymore.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s